He et al. (2026) The general formulation for mean annual runoff components estimation and their change attribution

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-02-02
• Authors: Yufen He, Changming Li, Hanbo Yang
• DOI: 10.5194/hess-30-553-2026

Research Groups

State Key Laboratory of Hydroscience and Engineering, Department of Hydraulic Engineering, Tsinghua University, Beijing, China.

Short Summary

This study proposes a general and concise Modified Ponce-Shetty (MPS) model to quantify and attribute mean annual surface flow, baseflow, and total runoff. Applied across 662 catchments in China and the contiguous United States, the model accurately depicts spatial variability and simulates multi-year runoff components, revealing that surface flow is jointly controlled by precipitation and environmental factors, while baseflow is primarily influenced by environmental factors.

Objective

• Develop a general and concise formulation (MPS model) to describe mean annual runoff components variability.
• Validate and compare the MPS model's performance against Budyko-type formulations.
• Attribute variations of runoff components to changes in precipitation and other environmental factors.

Study Configuration

• Spatial Scale: 662 catchments (312 in China, 350 in the contiguous United States).
• Temporal Scale: Mean annual time scale, derived from daily hydrological and meteorological data.
  • China: 1960–2000 (calibration: 1960–1990; validation: 1991–2000).
  • Contiguous United States (CONUS): 1980–2014 (calibration: 1980–2000; validation: 2001–2014).

Methodology and Data

• Models used:
  • Modified Ponce-Shetty (MPS) model (proposed).
  • Original Ponce-Shetty model (for approximation evaluation).
  • Budyko-type formulations (for comparison).
  • One-parameter Lyne-Hollick digital filter (for baseflow separation).
  • Penman equation (for potential evapotranspiration estimation).
• Data sources:
  • China: China Gauge-based Daily Precipitation Analysis (CGDPA) (0.25° spatial resolution precipitation), China Meteorological Data Service Center (meteorological data), Hydrological Bureau of the Ministry of Water Resources of China (discharge data).
  • CONUS: CAMELS dataset (daily precipitation, observed runoff, aridity index).

Main Results

• The MPS model effectively describes the spatial variability of runoff components across catchments:
  • China: R² of 0.86 for surface flow (Qs), 0.69 for baseflow (Qb), and 0.91 for total runoff (Q).
  • CONUS: R² of 0.81 for Qs, 0.44 for Qb, and 0.80 for Q.
• The model accurately simulates multi-year runoff components in validation periods, with R² exceeding 0.97 and root mean square error (RMSE) less than 66 mm.
• The MPS model also effectively estimates surface flow fraction (SFC) and baseflow fraction (BFC) with R² exceeding 0.94 and RMSE of 0.03.
• Spatial patterns of model parameters (wetting potential Wp, vaporization potential Vp, evapotranspiration potential Up) are consistent with climate zoning, showing smaller values in humid regions and larger values in arid regions.
• The MPS model accurately quantifies changes in runoff components between two long-term periods:
  • Surface flow change (ΔQs) attribution: R² = 0.99, RMSE = 1.6 mm.
  • Baseflow change (ΔQb) attribution: R² = 0.90, RMSE = 16 mm.
  • Total runoff change (ΔQ) attribution: R² = 0.91, RMSE = 42 mm (outperforming the Budyko framework).
• Attribution analysis indicates that surface flow variation is jointly controlled by precipitation and environmental factors, while baseflow variation is mainly influenced by vaporization potential (Vp) in most catchments.

Contributions

• Development of a general and concise formulation (MPS model) for quantifying and attributing mean annual surface flow, baseflow, and total runoff, unifying different runoff components within a single framework.
• The MPS model characterizes runoff components as a function of available water with only one parameter, simplifying complex parameterizations compared to existing Budyko-type formulations and the original Ponce-Shetty model.
• Demonstrated superior performance in estimating baseflow (Qb) and baseflow fraction (BFC) compared to Budyko-type formulations.
• Provided novel insights into the long-term catchment water balance and the distinct controlling factors for surface flow (precipitation and environmental factors) and baseflow (environmental factors).
• Offers an efficient tool for large-sample hydrological investigations and exploring future runoff variations under changing climate.

Funding

• National Natural Science Foundation of China (grant nos. 42041004, 52309022)
• National Key Research and Development Program of China (grant nos. 2021YFC3000202, 2022YFC3002802)

Citation

@article{He2026general,
  author = {He, Yufen and Li, Changming and Yang, Hanbo},
  title = {The general formulation for mean annual runoff components estimation and their change attribution},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-553-2026},
  url = {https://doi.org/10.5194/hess-30-553-2026}
}